What's new
Fantasy Football - Footballguys Forums

Welcome to Our Forums. Once you've registered and logged in, you're primed to talk football, among other topics, with the sharpest and most experienced fantasy players on the internet.

Value Chart for CFB? (1 Viewer)

I wish I could give a solid date. The answer is: If everything goes well, a day or two after projections are done, which should be a day or two after game slates and player salaries are available. (FanDuel has posted only its Thursday night 9/3 contests, not its Saturday 9/5 contests. DraftKings just posted their Saturday contests within the last day or two, I think.)

 
Last edited by a moderator:
Does anyone know what the average number of points you are looking to score to win for both cash and tournament CFB games? For NFL its 120 and 180, is there data on CFB or am I just missing it?

 
I've got the college version of the IVC working (I think). We're just finishing up the projections to load into it.

 
Here's the chart for the Thursday night and early Saturday games at DraftKings.

http://subscribers.footballguys.com/apps/article.php?article=tremblay_CFB15DraftKings01

(There are more QBs listed than RBs or WRs because we have projections for every QB, but projections only for RBs and WRs who are in the top 125 at their position, and some of the Week 1 slate includes teams with guys outside that range.)

FanDuel should be up soon as well.

 
Last edited by a moderator:
Awesome product! However, I do notice that the projected points listed in the interactive chart do not match the projected DK points from Devin's posted projections. The value chart projects AJ Oullette for 31.9 points and Devins projections say 29.8 points just to give one example.

 
Moss1116 said:
Awesome product! However, I do notice that the projected points listed in the interactive chart do not match the projected DK points from Devin's posted projections. The value chart projects AJ Oullette for 31.9 points and Devins projections say 29.8 points just to give one example.
Click to expand...
Let me ask Devin about that ...

 
Devin's on a plane, but I imagine the projected points on the site are more current than the ones I had, so the IVC is now using the ones loaded on the site.

 
Last edited by a moderator:
Interesting that James Washington is projected higher than Brandon Sheperd, any additional insight into those projections?

 
I don't know Devin's thoughts, but Sheperd is coming back from a dislocated finger and Washington is probably the more dangerous deep threat. If his finger is okay, Sheperd does seem like the more reliable option...

 
Hey sorry guys, and thanks Maurile for covering for me. I can definitely help answer any projections questions, and will be having a much larger influence over the course of the college football season.

Regarding the discrepancy between the IVC and the projections, yes this was just a timing issue that I made a few minor tweaks to the projection model. This is something that won't happen again, as this week has been a little bit crazy trying to get all the projections finalized for the first time.

Regarding Oklahoma State, this is one of the most asked questions that I have received. The answer is unfortunately not as simple as a dislocated finger. Two things, first the model is backloaded meaning there is more weight put on the end of last year than the beginning. The model also discounts the outlier games, so it is going to downgrade a little bit the guy who has one breakout game such as Sheperd. It also puts a larger improvement factor on Freshman going into their sophmore season (Washington) than it does a junior going into his senior season. This has been something that I have tracked over the last few years and freshman have the biggest improvement going into their second year, most of it is because of increased play counts, as well as increased role in the offense/more trust with their coach.

On a week to week basis they're going to be close, the model just slightly grades Washington out a little bit higher for the factors above.

 
Thanks for the reply, Devin. I wasn't aware that it was a model's projections rather than a individual's. That actually makes me like it more, as I fully understand the limitations of models and how they will sometimes have slightly different-than-expected results. I would definitely be interested in reading your thoughts on the model's output each week in addition to the actual data, so its awesome that you're willing to answer questions in this thread.

 
I plan on writing a full breakdown on the projections model that we use at some point once this week is done to give you guys an in-depth look at how we pull together the process, as well as the manual adjustments we sometimes have to make.

 
do you guys have weekly statline projections anywhere? I know you have projected fantasy points in the value chart, but is there also a plan to release individual stat lines? (completions/attempts, receptions, yards, TDs).

if not, can there be?

 
New to the site, but wanted to start off by saying how impressed I am. So many great resources, including this fantastic tool!

I'm not sure if I'm missing something, but it doesn't seem to me that there is any Friday information on the FD IVC. One of the options says Thurs-Fri, but it looks like strictly Thursday to me. Is there any plan to add Friday information?

 
How often is this chart update? The Early Sat chart was just now updated with the early games. It previously just had after 1pm

EDIT* users should verify the games included in the Value Chart match what is being offered by DraftKings tourneys, because they did not this morning for the Sat Early contests

 
Last edited by a moderator:
It would be nice if the IVC for college football noted the date and time of the last update to the projections like it does for NFL.

 
SwampDawg said:
Any idea what the regular release schedule for these is going to be ?
Click to expand...
I'm not sure. We're still finding our rhythm, and so are FanDuel and DraftKings. They've finally released early Saturday game slates (along with Thursday). The CFB crew is getting their projections done and then the IVCs will go up just after that. But I honestly don't have an ETA for when all of that will regularly occur in future weeks. It will help greatly if the hosting sites release their game slates earlier.

 
Last edited by a moderator:
Rock Lonemilk said:
It would be nice if the IVC for college football noted the date and time of the last update to the projections like it does for NFL.
Click to expand...
I can do that. I commented that part of the code out during Week 1 because I didn't know whether projections would be updated after initial posting, but I'll restore it for this week.

 
Last edited by a moderator:
Looks like we'll have the Thursday and early Saturday slates posted late tonight. The late Saturday slates still haven't been released yet by FD and DK, so I don't know when we'll have those.

 
Sorry for the non-response guys, have been busy working on the projections, and finally get a chance to come on here and address some of the questions.

This week was a little bit strange due to the Sunday/Monday games causing the Thursday slates to be delayed. Going forward we expect the projections to be done two days prior to the game start time assuming the slates are announced at that time. The only issue may be at times with the Saturday night slate as these are sometimes released later, but the sites did a great job this week releasing the slates in a timely fashion. The Saturday night slate was released around 7PM CST and we had the projections finalized by around 1AM CST.

Timeline Goal in Normal Week

Thursday/Friday Slate- Tuesday Release

Saturday Early Slate- Thursday Release

Just to give some of you guys a little insight into our projections, we create projections for all 128 FBS teams on Sunday and Monday which is a total of approximately 600 players. As the season continues to progress, we will be looking to continue to add additional players to the projections. The initial load of a player is the only manual piece of this process as we establish a "baseline" for each player individually. This baseline while it is a calculation has some manual things included in it. Since there is so much turnover in college football, the baseline looks at the players previous stats if he was in the same role, and the growth rate of a college player (Freshman to Sophmore sees a bigger lift than Junior to Senior), and the historical statistics of that position under that coach/school. The manual adjustment comes in regarding things such as skills adjustments.

One example that I use a lot is Corey Clement. If we were to just look at the past running backs at Wisconsin, we would see names such as Melvin Gordon, Montee Ball, James White. Therefore, if we were to project Corey Clement, without adjusting it, he would likely rank out as the top running back solely on the merit of the previous running backs who came before him. While we like Corey Clement quite a bit this year, he is not the number one running back. The other time we will make an adjustment is losses to the offensive line, so if a running back had 5 seniors on the offensive line last year, and now had a group of Freshman/Sophomores, the model isn't smart enough yet to apply a value to this, so I will go in and make an adjustment. In college football, so many things change on a week to week basis, which is why I watch around 30 college football games per week in order to make sure there are any adjustments that need to take place. While this takes a significant amount of time, I have found that this is the best approach.

Feel free to reach out to me at any time with any questions and I'll do my best to answer

 
Quick question about the Draftkings chart. I'm taking my first dip in the CFB pool. I know zero about CFB but based on some comments by Maurile in another thread I'm going to try it out on a very small scale. I am relying 100% on the interactive chart to try out some small cash games.

My problem so far has been lining up contests I want to enter with the players actually listed in the chart. I have tinkered around with the tabs for early, late, night, etc. I get that different contests are for different time blocks. However, I still can't seem to easily and quickly assemble a lineup for a given contest as many of the players suggested by the chart are not even in the Draftkings slate of players. They have either played already or are not in the Draftkings list at all. I tried manually removing players from the chart but even that seemed to be a dead end and was taking too long given that I have no idea who has played, who plays when and who these guys even are.

Am I doing something wrong? This is definitely not a complaint about the chart as it is awesome. Just wondering if there is a quick solution or does it just require a bit more leg work on my end. Thanks!

 
steviey said:
Quick question about the Draftkings chart. I'm taking my first dip in the CFB pool. I know zero about CFB but based on some comments by Maurile in another thread I'm going to try it out on a very small scale. I am relying 100% on the interactive chart to try out some small cash games.

My problem so far has been lining up contests I want to enter with the players actually listed in the chart. I have tinkered around with the tabs for early, late, night, etc. I get that different contests are for different time blocks. However, I still can't seem to easily and quickly assemble a lineup for a given contest as many of the players suggested by the chart are not even in the Draftkings slate of players. They have either played already or are not in the Draftkings list at all. I tried manually removing players from the chart but even that seemed to be a dead end and was taking too long given that I have no idea who has played, who plays when and who these guys even are.

Am I doing something wrong? This is definitely not a complaint about the chart as it is awesome. Just wondering if there is a quick solution or does it just require a bit more leg work on my end. Thanks!
Click to expand...
Yeah definitely don't rely on the chart alone. It's good but there's always a couple random outliers at the top of the value or h-value list that don't belong. What I've been doing and have been successful with is combining the value chart with the slate breakdowns.

 
robert88 said:
steviey said:
Quick question about the Draftkings chart. I'm taking my first dip in the CFB pool. I know zero about CFB but based on some comments by Maurile in another thread I'm going to try it out on a very small scale. I am relying 100% on the interactive chart to try out some small cash games.

My problem so far has been lining up contests I want to enter with the players actually listed in the chart. I have tinkered around with the tabs for early, late, night, etc. I get that different contests are for different time blocks. However, I still can't seem to easily and quickly assemble a lineup for a given contest as many of the players suggested by the chart are not even in the Draftkings slate of players. They have either played already or are not in the Draftkings list at all. I tried manually removing players from the chart but even that seemed to be a dead end and was taking too long given that I have no idea who has played, who plays when and who these guys even are.

Am I doing something wrong? This is definitely not a complaint about the chart as it is awesome. Just wondering if there is a quick solution or does it just require a bit more leg work on my end. Thanks!
Click to expand...
Yeah definitely don't rely on the chart alone. It's good but there's always a couple random outliers at the top of the value or h-value list that don't belong. What I've been doing and have been successful with is combining the value chart with the slate breakdowns.
Click to expand...
Completely agree with this approach, which is why we are doing slate breakdowns. From a projections standpoint, we created projections for 650 players across all FBS teams. This is an increase of about 300 from last week, and as we continue to have data there will be additional players added. The challenge is from a projections standpoint creating projections for players who have an unclear role in the offense.

As of next week we hope to have around 800 players total, and once we get to conference play starting we should have over 1,000 players in our projections model.

I am going to spend a little bit more time looking through this to look into any of the other issues.

Thanks Guys

 
steviey said:
My problem so far has been lining up contests I want to enter with the players actually listed in the chart. I have tinkered around with the tabs for early, late, night, etc. I get that different contests are for different time blocks. However, I still can't seem to easily and quickly assemble a lineup for a given contest as many of the players suggested by the chart are not even in the Draftkings slate of players. They have either played already or are not in the Draftkings list at all. I tried manually removing players from the chart but even that seemed to be a dead end and was taking too long given that I have no idea who has played, who plays when and who these guys even are.

Am I doing something wrong? This is definitely not a complaint about the chart as it is awesome. Just wondering if there is a quick solution or does it just require a bit more leg work on my end. Thanks!
Click to expand...
First, make sure you are looking at the Week 2 chart, not the Week 1 chart that was linked earlier in this thread (last week).

For the Week 2 chart, we had four game slates featured. The Thurs-Fri slate and the Early Sat slate have already started, so they're no longer relevant.

The Sat Afternoon slate corresponds to the 12-game slate starting at 5pm, and the Late Sat slate corresponds to the 12-game slate starting at 8pm.

The IVC does not have the two other slates that DK is currently offering -- the 5-game slate starting at 9:15 pm and the 4-game slate starting at 10:15pm.

The players in the IVC slates should line up exactly with the players in those DK slates. If you see some that aren't, please let me know, either by posting here or by emailing me at tremblay@footballguys.com.

 
Yeah, this looks to be a short experiment with an inactive guy in my first two lineups. Does FBGs have a Saturday morning inactive list like they do for Sunday morning? I can't locate one.

 
robert88 said:
Yeah definitely don't rely on the chart alone. It's good but there's always a couple random outliers at the top of the value or h-value list that don't belong. What I've been doing and have been successful with is combining the value chart with the slate breakdowns.
Click to expand...
Ricky Jones and Jordan Howard were outliers earlier in the week. We didn't notice that there were two players with each of those names, and the IVC was combining them (using the salary for one Ricky Jones and the projection for the other Ricky Jones) making him look like a great bargain even though he wasn't. Same with Howard.

That screwed up my own lineups in the early slate, but the same-name issue won't happen again.

 
i've made 2 lineups using the interactive value chart and both scored drastically below expectations. am i doing it wrong? whats going on here?

 
wow and for tonights late games i used the texas qb because the chart told me to and hes not playing. awesome! thanks guys!

 
Ya I did the same, although I'm certainly also to blame considering how little I know about college football. Is there a good source for depth charts for CFB? Like did everyone else know that Swoopes wouldn't be starting?

 
Ive been checking everything against ourlads.com

I am very confidant in saying the value chart is extremely helpful. I was about 2 to 1 in my college test games last week. The miss price of a player actually cost me quite a bit of that profit too, as I rushed unchecked lineups into contests with large overlay and still didnt cash. This week in closer to 100 contests Im going to cash everything except some of my GPPs. I hit twelfth out of 5300 earlier today and I am currently sitting at 27th at the end of another. Just realize the computer is a mindless algorithm, and if your own research is telling you one of the guys it recommends isnt going to see the field, then go with someone else.

 
Thanks Nixon, that looks like a great resource. I'll be using it from now on. Interestingly, it has Swoopes as the first string QB for Texas. So maybe want happened tonight was difficult to predict.

 
NixonMask said:
http://collegefootball.ap.org/article/freshman-qb-heard-propels-texas-over-rice-42-28

If Ive learned one thing from this season so far it is to stay away from the dumpster fires, even more so than the NFL, because theres so much volatility. With so many teams there is always going to be a reliable, if not always exciting option somewhere on the slate.
Click to expand...
It really depends on the dumpster fire and the players involved. College football is so difficult because you really have to understand the system and the mentality of of the head coach.

 
Thanks a lot for your help guys. The fact that Heard didn't even know he was starting until immediately before the game makes me feel better about my decision, but worse because the same thing could happen again in the future. The resources you guys supplied look like very good ones, so hopefully that will help moving forward. Thanks again!

 
Can I get one of the Football guys opinions on Skov this week? His price is super deflated from all the blowouts he hass been off the field for. Is his H-value a reflection of the computer using those stats, or do you not like him this week even at that price?

 
NixonMask said:
Can I get one of the Football guys opinions on Skov this week? His price is super deflated from all the blowouts he has been off the field for. Is his H-value a reflection of the computer using those stats, or do you not like him this week even at that price?
Click to expand...
Here's the problem with Skov. He is likely only to get 15 carries and is facing a great defensive lineman in Sheldon Day and the best middle linebacker in recent college football memory in Jaylon Smith.

Skov is the B Back in the triple option offense which essentially means most of his runs are going to be right up the middle targeting the middle linebacker in the option offense. The problem with this position is that he won't catch passes and doesn't have the upside of a traditional running back, so you're essentially relying on a touchdown.

With Draftkings pricing being as tight as it is, $4,900 is not a horrible price, but he is not a fantastic play as you're essentially hoping for a short yardage touchdown. He does however have GPP upside due to his ability to get 2-3 short yardage touchdowns.

 
You must log in or register to reply here.

Similar threads

Steeler
No Vegas Value Chart this year for DFS?
Replies
3
Views
339
Joe Bryant
Joe Bryant
SwampDawg
Draftkings Core 4
Replies
33
Views
2K
TheBigHouse
TheBigHouse
E
Chalk RB and WR….taste good this week? Are you eating it?
Replies
1
Views
336
BassNBrew
BassNBrew

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Top